Crossplane: Infrastructure Automation Ledger 2026 Scorecard
Crossplane makes a Kubernetes cluster a provisioning control plane that repairs drift continuously; v2 in 2025 rebuilt the model and the CNCF graduated it.
Where Crossplane put its effort, 2019 to 2026
Each square is one capability area in one year. The brighter it is, the more the area gained in the Infrastructure Automation Ledger that year; 1, 2 and 3 mark that year's biggest pushes.
| ’19 | ’20 | ’21 | ’22 | ’23 | ’24 | ’25 | ’26 | |
|---|---|---|---|---|---|---|---|---|
| Provisioning breadth | ||||||||
| Configuration management depth | 1 | |||||||
| State, planning and safety | 2 | 1 | ||||||
| Language and authoring | 1 | 2 | ||||||
| Orchestration and self-service | 1 | 3 | 2 | |||||
| Scale and control plane | 3 | |||||||
| Security and compliance | 2 | 3 | ||||||
| Bare metal and day-2 operations | ||||||||
| Ecosystem and integrations | 1 | 3 | ||||||
| AI assistance |
Biggest pushes overall: Orchestration and self-service (+3), Language and authoring (+2.5), Security and compliance (+2.5). Already strong in 2018: Provisioning breadth, State, planning and safety, Language and authoring.
Top focus each year: 2020 Orchestration and self-service; 2021 Ecosystem and integrations; 2024 Language and authoring; 2025 Configuration management depth; 2026 State, planning and safety. 3 of 8 years show no gain in any area.
Part of the Infrastructure Automation Ledger (edition 2026-10-16). How the scores work: methodology.
Snapshot
- Type: community project (CNCF Graduated), created and primarily maintained by Upbound
- Owner: Cloud Native Computing Foundation; CNCF Sandbox 2020-06-25, Incubating 2021-09-14, Graduated 2025-10-28 (announced 2025-11-06). Upbound, Inc. (Seattle, venture backed, $69 million raised) is the creator and primary maintainer
- Engine: the project is the engine; Upbound Crossplane (UXP) is a drop-in distribution
- Licence: Apache 2.0 with no change found. Upbound editions (2026-10-15): Community free, Standard from $1,000 a month (up to 5 control planes), Enterprise and Business Critical custom; self-hosted Space only on Business Critical
- Latest release: v2.4.2 (2026-09-22); v2.4.0 (2026-08-20); v1.20.13 (2026-09-15) on the last v1 series
- Support lifecycle: A release every 13 weeks; the three newest minors maintained for nine months each, best effort with no response SLA; v2.4 to May 2027, v2.3 to February 2027, v2.2 and v1.20 to November 2026; upgrade to v2 from v1.20
- Next milestone: v2.5 in November 2026 (published calendar runs to v2.7 in May 2027)
- Targets: vSphere: stuttgart-things/provider-vspherevm, community, Upjet-generated, v0.8.1 (2026-06-13). Proxmox VE: valkiriaaquatica/provider-proxmox-bpg, community, v1.21.0 (2026-10-09). Nutanix: mgeorge67701/provider-nutanix, community, v0.2.9 on the Upbound Marketplace, release date not found. Hyper-V and XCP-ng: no provider found. KubeVirt: no dedicated provider; since v2.0 a composition can include any Kubernetes resource, including VirtualMachine objects. Clouds through Upbound Official Providers (gated) and crossplane-contrib
- Tier: Infrastructure-as-code engine
- Archetype: Dependable builder
- Evidence grade: B. The release cycle page, the README release table, the v1.0, v1.17, v1.20, v2.0 and v2.4 release notes, the licence, the CNCF project page and Upbound's provider policy posts were read as primary sources. The contents of v2.1 to v2.3 were read only in part, and the years of pre-2025 releases are inferred from release pages that show day and month only (grade B). Upbound's AI claims and the first release date are unverified (grade C).
Scope of the scores: The scores cover the open source engine as a community project. Upbound's commercial distribution (UXP, Spaces, Marketplace) is described where it changes what a user can download; since 2025-03-25 Upbound's Official Providers are free only in their latest version and, from UXP 2.0, run only on UXP.
Scorecard
| Dimension | Score | Direction |
|---|---|---|
| Capability today | 50.1 / 100 | higher is better |
| Momentum since end of 2022 | +17.5 points (2018 to 2022: +18.2) | higher is better |
| Credibility | 20.0 / 25 | higher is better |
| Sustainability risk | 9.0 / 25 | lower is better |
| Technical lock-in | 11.5 / 25 | lower is better |
| Integrator fit | 13.0 / 20 | higher is better |
| AI leverage | 2.0 / 20 | higher is better |
| Ledger Index | 57.2 / 100 | balanced weights; see the methodology |
Capability by domain, 2015 to today
Scores 0 to 5 against fixed anchors; total is weighted to 100.
| Domain (weight) | 2015 | 2019 | 2022 | Today |
|---|---|---|---|---|
| Provisioning breadth (15) | n/a | 1 | 2 | 2.5 |
| Configuration management depth (12) | n/a | 0 | 0 | 1 |
| State, planning and safety (12) | n/a | 1 | 2 | 3 |
| Language and authoring (10) | n/a | 1 | 2 | 3.5 |
| Orchestration and self-service (10) | n/a | 1 | 2.5 | 4 |
| Scale and control plane (8) | n/a | 1 | 2.5 | 3.5 |
| Security and compliance (10) | n/a | 0.5 | 2 | 3 |
| Bare metal and day-2 operations (8) | n/a | 0.5 | 0.5 | 0.5 |
| Ecosystem and integrations (8) | n/a | 1 | 2.5 | 3.5 |
| AI assistance (7) | n/a | 0 | 0 | 0 |
| Total / 100 | n/a | 14.4 | 32.6 | 50.1 |
Year by year: 2018: 14.4 · 2019: 14.4 · 2020: 31.8 · 2021: 32.6 · 2022: 32.6 · 2023: 32.6 · 2024: 36.8 · 2025: 46.1 · 2026: 50.1
What moved the score
| Year | Domain | Change | Points | Trigger |
|---|---|---|---|---|
| 2020 | Ecosystem and integrations | 1 → 2 | +1.6 | CNCF Sandbox (2020-06-25) |
| 2020 | Language and authoring | 1 → 2 | +2.0 | 1.0: XRDs and Compositions as v1 APIs |
| 2020 | Orchestration and self-service | 1 → 2.5 | +3.0 | 1.0: Compositions as the self-service model behind a Kubernetes API |
| 2020 | Provisioning breadth | 1 → 2 | +3.0 | 1.0: provider APIs declared v1 with a stability commitment |
| 2020 | Scale and control plane | 1 → 2.5 | +2.4 | 1.0 on a highly available Kubernetes control plane |
| 2020 | Security and compliance | 0.5 → 2 | +3.0 | 1.0: Kubernetes RBAC and audit apply to every object |
| 2020 | State, planning and safety | 1 → 2 | +2.4 | 1.0: continuous reconciliation of managed resources repairs drift |
| 2021 | Ecosystem and integrations | 2 → 2.5 | +0.8 | CNCF Incubating (2021-09-14) |
| 2024 | Language and authoring | 2 → 3 | +2.0 | 1.17: composition functions v1, native patch and transform deprecated |
| 2024 | Orchestration and self-service | 2.5 → 3 | +1.0 | 1.17: function pipelines GA |
| 2024 | State, planning and safety | 2 → 2.5 | +1.2 | 1.17: render and xpkg init GA for local checks |
| 2025 | Configuration management depth | 0 → 1 | +2.4 | 2.0: Operation type for one-off, scheduled and event-driven tasks; no OS configuration |
| 2025 | Ecosystem and integrations | 2.5 → 3.5 | +1.6 | CNCF Graduated (2025-10-28); 1.20: default registry xpkg.crossplane.io; 2.0 composes any Kubernetes resource |
| 2025 | Orchestration and self-service | 3 → 4 | +2.0 | 2.0: namespaced composite resources; Operation type for scheduled and event-driven tasks |
| 2025 | Provisioning breadth | 2 → 2.5 | +1.5 | 2.0: compositions can include any Kubernetes resource; community providers for vSphere, Proxmox VE and Nutanix |
| 2025 | Scale and control plane | 2.5 → 3 | +0.8 | 2.0: managed resource filtering limits installed CRDs |
| 2025 | Security and compliance | 2 → 2.5 | +1.0 | CNCF graduation: two security audits; signed packages and SBOMs for official packages |
| 2026 | Language and authoring | 3 → 3.5 | +1.0 | 2.3: render against the real reconciler as test tooling; 2.2: pipeline inspector (alpha) |
| 2026 | Scale and control plane | 3 → 3.5 | +0.8 | 2.4: safe-start providers scale to zero; crossplane-runtime 2.4 client caching cuts provider memory (vendor figure 80 percent) |
| 2026 | Security and compliance | 2.5 → 3 | +1.0 | 2.4: garbage collector deletes only resources owned by the composite; scannable images |
| 2026 | State, planning and safety | 2.5 → 3 | +1.2 | 2.3: render runs the real reconciler; 2.4: realtime reconciliation when required resources change; no plan or diff preview |
Direction, 2023 to 2026
Each release item carries one theme and one driver (V vision, C customer need, M market window, U upstream, P portfolio).
| Date | Item | Theme | Driver |
|---|---|---|---|
| 2023-01 | 1.11: composition functions, alpha | Language | V |
| 2023-09 | Upbound contributes control plane provider technology to Crossplane under the CNCF | Ecosystem | P |
| 2023-11 | 1.14: composition functions, beta | Language | V |
| 2024-08 | 1.17: composition functions v1, render and xpkg init GA, native patch and transform deprecated | Language | V |
| 2025-01 | Upbound Official Providers: free only in the latest version from 2025-03-25 | Packaging | P |
| 2025-05 | 1.20: default registry xpkg.crossplane.io, realtime compositions beta | Ecosystem | C |
| 2025-08 | 2.0: namespaced composite and managed resources, compositions of any Kubernetes resource, Operations | Orchestration | C |
| 2025-08 | UXP 2.0: Official Providers run only on Upbound's distribution | Packaging | P |
| 2025-11 | CNCF graduation announced | Ecosystem | V |
| 2026-02 | 2.2: ImageConfig for packages, pipeline inspector alpha | Language | C |
| 2026-03 | function-kro: YAML plus CEL composition | Language | V |
| 2026-05 | 2.3: render runs the real reconciler, provider deletion protection alpha | State and planning | C |
| 2026-08 | 2.4: realtime reconciliation on required resources, scale-to-zero providers, garbage collector hardening, CLI renamed | Scale | C |
| 2026-08 | Upbound v3: Hub, Console, fleet-wide identity, self-hosted and air-gapped platform | Scale | P |
| 2026-10 | New Crossplane CLI developer experience announced; crossplane-runtime 2.4 memory reduction | Language | C |
- Centre of gravity: Language (6), Ecosystem (3), Packaging (2)
- Blind spots (no items): Provisioning breadth, Configuration management, Security, Bare metal, AI, Licensing
- Driver mix: Vision 33%, Customer need 40%, Market window 0%, Upstream 0%, Portfolio 27%
- Idea-to-need index: 0.45 (0 = demand-pulled, 1 = thesis-pushed)
Credibility: 20.0 / 25
| Criterion | Score | Basis |
|---|---|---|
| Cadence | 5 | Published 13-week cycle with a calendar to May 2027; v2.2, v2.3 and v2.4 shipped on their dates |
| Roadmap transparency | 4 | Public release calendar, open repository and release notes; the feature roadmap was not read |
| Say-do | 4 | Release dates held through 2026; composition functions went alpha to v1 in 19 months; Upbound's AI items have no stated status |
| Velocity | 4 | Several capabilities in 2.0 and 2.4; 2.2 and 2.3 content read only in part |
| Lifecycle stability | 3 | Nine months per minor, three supported; v1 to v2 upgrade only from v1.20; CLI renamed in 2.4; native patch and transform deprecated in 1.17 |
Sustainability risk: 9.0 / 25 (lower is better)
| Criterion | Score | Basis |
|---|---|---|
| Maintainer concentration | 3.5 | Upbound is the creator and primary maintainer and the graduation post names no other maintainer organisation; 3,000+ contributors from 480+ companies are vendor-reported |
| Backer health | 2 | Upbound is venture backed with $69 million raised; the CNCF hosts the project; no sponsor list read |
| Licence-change risk | 1.5 | Apache 2.0 held by the CNCF; the risk sits in the providers, where Upbound gated Official Providers in 2025 and tied them to UXP |
| Abandonment signs | 0.5 | Quarterly releases kept, calendar published to 2027, CNCF Graduated in 2025 |
| Security response | 1.5 | Two security audits before graduation; 2.4 hardened garbage collection; maintenance is best effort with no response SLA |
Technical lock-in: 11.5 / 25 (lower is better)
| Criterion | Score | Basis |
|---|---|---|
| Formats | 2.5 | Kubernetes YAML, but XRDs, Compositions and functions are Crossplane specific |
| Export path | 3 | No documented path to another IaC tool; managed resources can be left in place when a claim is deleted |
| Stack coupling | 3 | Requires a Kubernetes control plane |
| Hardware / cloud coupling | 0.5 | Any cluster, offline capable; the engine needs no Upbound account |
| Skills and tooling coupling | 2.5 | Kubernetes skills are common; composition and function authoring skills are scarce |
Integrator fit: 13.0 / 20
| Criterion | Score | Basis |
|---|---|---|
| Partner programme openness | 2.5 | Foundation project with open downloads; Upbound's partner terms not found and its Official Providers are gated |
| Multi-tenancy and self-service | 4 | Namespaced composite resources in v2 give per-tenant APIs under Kubernetes RBAC |
| API and infrastructure-as-code | 4.5 | The Kubernetes API is the whole interface; any client and any GitOps tool works |
| Skills and certification | 2 | No certification found |
AI leverage: 2.0 / 20
How the product uses AI, scored per year like capability. Only GA counts.
| Criterion | 2019 | 2022 | Today | Latest step |
|---|---|---|---|---|
| AI code and plan generation | 0 | 0 | 0 | 2018: None |
| AI platform stack | 0 | 0 | 0 | 2018: None |
| AI-assisted operations | 0 | 0 | 0 | 2018: None |
| Agent openness | 1 | 2 | 2 | 2020: 1.0: every object behind a stable Kubernetes API that any agent can drive; no official MCP server verified |
Year by year: 2018: 1 · 2019: 1 · 2020: 2 · 2021: 2 · 2022: 2 · 2023: 2 · 2024: 2 · 2025: 2 · 2026: 2
AI say-do, 2023 to 2026: 2 AI items announced; 0 GA, 0 still preview, 2 pending, 0 dropped. UXP 2.0's AI-enhanced operations and the Upbound Marketplace MCP server were announced without a stated status; neither was read from a primary page.
The project's own position is that a Kubernetes API is the agent interface, and it ships no AI feature. Upbound announces AI insights and an MCP server for its distribution, but no status, date or documentation page could be verified in this pass.
Reading
Crossplane scores highest in this group on orchestration and self-service because the product is a Kubernetes API that platform teams expose to tenants, and it is the only one that repairs drift continuously rather than reporting it. It scores lower on state and planning because there is no plan or diff before a change reaches the cluster, and on breadth because the on-prem hypervisor providers are maintained by individuals. The cadence record is the best in the group, with dates published two years ahead and kept. Check your dependence on Upbound before adopting: the Official Providers for the big clouds are free only in their latest version and, from UXP 2.0, run only on Upbound's distribution.
Open questions and items to re-score
- The first release date and the exact years of pre-2025 releases (GitHub pages show day and month only).
- Full contents of v2.1 to v2.3.
- Whether the community vSphere, Proxmox VE and Nutanix providers have more than one maintainer.
- UXP AI features and the Upbound Marketplace MCP server: status, date and documentation.
- Upbound Official Provider list prices and partner programme.
- Pipeline inspector and provider deletion protection are alpha; re-score when GA.
Related reading
More from Infrastructure Automation Ledger 2026
- Overview: all platforms side by side
- Infrastructure Automation Ledger Methodology: How Every Score Is Built
- Ansible AAP: Infrastructure Automation Ledger 2026 Scorecard
- Ansible: Infrastructure Automation Ledger 2026 Scorecard
- Atlantis: Infrastructure Automation Ledger 2026 Scorecard
- Bicep: Infrastructure Automation Ledger 2026 Scorecard
- CFEngine: Infrastructure Automation Ledger 2026 Scorecard
- Chef: Infrastructure Automation Ledger 2026 Scorecard
- Cinc: Infrastructure Automation Ledger 2026 Scorecard
- CloudFormation: Infrastructure Automation Ledger Scorecard
- Digger: Infrastructure Automation Ledger 2026 Scorecard
- env zero: Infrastructure Automation Ledger 2026 Scorecard
- Foreman: Infrastructure Automation Ledger 2026 Scorecard
- Harness: Infrastructure Automation Ledger 2026 Scorecard
- Ironic: Infrastructure Automation Ledger 2026 Scorecard
- Juju: Infrastructure Automation Ledger 2026 Scorecard
- MAAS: Infrastructure Automation Ledger 2026 Scorecard
- Metal3: Infrastructure Automation Ledger 2026 Scorecard
- OpenTofu: Infrastructure Automation Ledger 2026 Scorecard
- OpenVox: Infrastructure Automation Ledger 2026 Scorecard
- Pulumi: Infrastructure Automation Ledger 2026 Scorecard
- Puppet: Infrastructure Automation Ledger 2026 Scorecard
- Salt: Infrastructure Automation Ledger 2026 Scorecard
- Satellite: Infrastructure Automation Ledger 2026 Scorecard
- Scalr: Infrastructure Automation Ledger 2026 Scorecard
- Semaphore: Infrastructure Automation Ledger 2026 Scorecard
- Spacelift: Infrastructure Automation Ledger 2026 Scorecard
- Terraform: Infrastructure Automation Ledger 2026 Scorecard
- Terragrunt: Infrastructure Automation Ledger 2026 Scorecard
- Tinkerbell: Infrastructure Automation Ledger 2026 Scorecard